Windows XP has been excelent for me. My XP box was rock solid until the video card died (the fan on the GPU seized up). Somehow that damaged the motherboard too.
I built a new system (learned my lesson: intel processor, intel chipset, everything on board), and I gave Linux a shot since I had nothing to loose... the risk was negligible. The polish isn't there, but it's alright.
In terms of my productivity, it's just as good as anything else, better even thanks to KDE. But I have to work to get stuff the way I want it. I can't just assume any given thing will work, and if I am sure it'll work, I can't assume I'll be able to make it work in a reasonable amount of time. For something like dual monitors, I'd need to be able to give it a weekend to feel comfortable.
I'll probably put Windows on this machine, but only so I can do.NET stuff and because I have an extra hard drive. Now that it's working, I see no reason to replace Linux.
It holds more water than most of SCO's arguments, but I don't think it will fly. They're giving it away, sure, but they're not just giving away the binaries. They're releasing the source code under a license that allows companies in direct competition with them to sell it in competing products. Including SCO.
It devalues SCO UNIX, but it also devalues Tru64, HP-UX, Solaris, IRIX, and MacOS X. Oh yeah, and IBM's own AIX. That's what happens when everyone starts standardizing on something. It would be negligent if IBM didn't persue a Linux strategy.
Dunno about anyone else, but Dell will certainly sell you one if you want it. IBM and others probably would too, but I don't feel like checking.
Didn't you watch the keynote? They compared a dual G5 to a dual Xeon (an x86 system).
"Apple (If you use Macs): DC 5 Apple (otherwise) : DC 15
Modifiers: Is switching to linux: -20 Is switching from linux: +15 Is going after Microsoft: -10 _____ vs. SCO : -20 Files a BS patent: +10 Is being investigated by the US government for anti-trust or Fraud: -5" Makes crappy computers that break a lot: +50
Okay... I have Apple, a DC 5 check. They're not really going to or from Linux. They make OpenFirmware computers and support many of the same APIs, so we'll give them the -20 anyway. They're pretty much the original anti-microsoft, so they get the -10 easy. Doing well so far...
Oh wait. My iBook has needed repairs 6 times in 8 months, accounting for something like 15% of the time I've owned it. I asked them to replace it, but they refused. +50.
I'll have to roll a natural 25 to make this check. I roll a 3. While I have nothing against their IP practices, or their stance towards Microsoft, their computers are too crappy for me to ever consider buying one again, even if it's the best tool for the job.
"However, I will grant that the average/. user knows what they're doing with their data far more than my average user does and is less likely to cause self-inflicted damage."
eh...
I know what I'm doing and I've found some pretty impressive ways to shoot myself in the foot. Right now the knowledge just makes me dangerous. I wouldn't let me anywhere near a production system.
FreeBSD 4.8 doesn't work with my Gigabyte GA-8IG1000MK's ethernet controller... it's supposed to support it with the fxp(4) driver, but it doesn't. I was sad. I don't think it's fixed in 4.9, at least the release notes don't mention it.
Gentoo is acceptable, but I had trouble with it too. "emerge e100" installed a module for the wrong kernel version and broke it completely. I had to add support manually. Easy (about 20 seconds in "make menuconfig"), but annoying. Pretty good otherwise.
I think I'd like to use FreeBSD just because I like BSD, but it's not quite as slick as OpenBSD. I don't think I'd like OpenBSD as a desktop system...
I saw one in Australia a while ago that could only function with the DOS drivers, even in Windows. Really, really cheap brand, I'd never even heard of it before. That was the only one I had seen.
I've seen some weird problems on LG drives though. Specific combinations of hardware have made mine unusable after waking from hiberation on an XP system. Plain ol' suspend doesn't bother it. That drive doesn't have any problems on Linux systems though. But then, I don't use Mandrake...
"Many people don't have the luxury of grabbing a P4 or Opteron as there whims demand. Supporting older hardware also does not slow an OS down. If done properly, it forces you to think about how to do things RIGHT so you don't eat unnecessary cpu cycles."
It's not just for speed. Multiple architechtures can break bad code early.
"By having multiple register sets within a single core, and tagging requests/results, you can avoid the complexity of SMP entirely, while producing the effect of having multiple processors."
No... keeping all the execution units of a CPU busy all the time is hard, and there's a limit to how much you can do. Hyperthreading is perhaps the more elegant solution, but it has its problems too. It tends to thrash the cache with certain algorithms, and the scalability is limited.
Adding a core increases instruction throughput a lot without adding (too much) complexity, and whatever you might think complexity makes things really tough. It's simply cheaper to throw more cores at the problem.
Another thing you don't consider is that signal propogation delays are a very real problem in todays CPUs. That's only going to get worse. Some way to spread out intependant operation must be used to keep things running fast.
RISC made a lot more sense when you couldn't fit a CPU onto a single die. You couldn't just throw transistors at it like you can now.
The internal architechture isn't and shouldn't be exposed to the outside. If it is, then compatability is broken each time they want to change it, and that simply is not an option. Compilers would never have a chance to get very good.
RISC binaries are a bit bigger, but the internal opcodes x86 chips use are absoloutely huge, like over a hundred bits per instruction. That would fill up the cache in no time, and memory latency and bandwidth would become a bigger problem than it is now.
"I don't think any of the free *nixes are better or worse as desktop systems."
Eh -- I dunno. Everything targets Linux. Takes a lot less work to get stuff working on it, especially like video cards and stuff. Also, consider the better interactivity on 2.6. Not appropriate for servers yet, but should be fine for a plain old dekstop. OpenBSD can't even take full advantage of a P4 HT.
A firewall system is a different story. Your basic scrap computer will do just fine, as long as it can fit 2 PCI NIC's. It's better for security if it's a seperate box from everything else, and it's cheaper if the computer that stays on all the time can do so on like 20 watts.
"Since I hear only good things about pf"
gah... you don't even want to know. After you've screwed around with it for a few hours it's easier than a hardware router with a web interface. Complex stuff can be done in one line, and while it requires thought, it has a much smaller tendency to bite you in the ass than with the other firewalls I've dealt with, because the language is smart and intuitive.
I love PF. It beats the crap out of netfilter/ipchains, IPF, and it's almost easier to use than a web interface on a hardware router.
Dunno how happy you'll be with it as a desktop system. My OpenBSD machine lives in the basement crawlspace, and I'm reasonably sure it doesn't have a monitor or keyboard at the moment. I don't know what it's like with a GUI.
Authors are free to use whatever license they choose and I am free to not use their software. That said, I find the GPL hypocritical. "This is Free Software, subject to the following conditions.".
It seems to be working, but the BSD license seems to be working too (Before anyone says BSD is dying, consider that things like Apache use a BSD-like license).
Sadly, BSD is getting soft too. Everything on my OpenBSD system keeps working without me having to get my hands dirty. FreeBSD is worse. Don't even get me started on OS X...
NetBSD still has promise though. It's practically unusable. It doesn't even have sudo installed by default. If you're in a masochistic mood, try it out.
OpenSSH, OpenSSL and Apache are all released under BSD-like licenses. By BSD-like, I mean you don't have to distribute the source of derivative works. Dunno about the other stuff, didn't feel like checking.:)
"And, they have to kill BSD as a loophole for source; otherwise, they're in deep doo-doo in court since it seems they didn't realize that what they took is from BSD."
That would be very, very difficult. The rights that SCO is trying to work with are the same ones USL (aka AT&T/Novell) tried to use in the BSD lawsuit, which USL settled when BSD code was found in UNIX without the copyright notice. In the settlement, all the BSD's had to go back to an older code base and rewrite a few files, but according to the settlement, after they did that they were clean.
If SCO goes after them, it'll be useless because they've all still got the clean code, but SCO's still using the dirty code.
As I understand it, the reason you can't do that is because 2.6.x breaks some 2.4.x stuff. Having two root partitions could mitigate that, because you'd have a place to put the different binaries and configuration files.
Any claim SCO had over BSD was explicitly given up when Novell settled the lawsuit with Berkley. If SCO did go after BSD, the BSD code that was incorporated into SysV without a copyright notice would make SCO vulnerable to a countersuit, just like it did in the early 90's.
I believe, though I can't prove it, that Apple was responsible for the memory that went bad. It was 3rd party, and after they replaced the motherboard my iBook would crash daily. Eventually the memory failed and I had to RMA it. Kingston was happy to replace it. The problems started immediately after I got the iBook back. Between the time I replaced the memory and the time the motherboard broke again, it was rock solid. I believe I hadn't rebooted once in the ~45 day period.
Slashdot Mirror
Windows XP has been excelent for me. My XP box was rock solid until the video card died (the fan on the GPU seized up). Somehow that damaged the motherboard too. I built a new system (learned my lesson: intel processor, intel chipset, everything on board), and I gave Linux a shot since I had nothing to loose... the risk was negligible. The polish isn't there, but it's alright. In terms of my productivity, it's just as good as anything else, better even thanks to KDE. But I have to work to get stuff the way I want it. I can't just assume any given thing will work, and if I am sure it'll work, I can't assume I'll be able to make it work in a reasonable amount of time. For something like dual monitors, I'd need to be able to give it a weekend to feel comfortable. I'll probably put Windows on this machine, but only so I can do .NET stuff and because I have an extra hard drive. Now that it's working, I see no reason to replace Linux.
The problem is, he's going to turn it into a soap box, and he might ultimately hurt IBM.
It holds more water than most of SCO's arguments, but I don't think it will fly. They're giving it away, sure, but they're not just giving away the binaries. They're releasing the source code under a license that allows companies in direct competition with them to sell it in competing products. Including SCO.
It devalues SCO UNIX, but it also devalues Tru64, HP-UX, Solaris, IRIX, and MacOS X. Oh yeah, and IBM's own AIX. That's what happens when everyone starts standardizing on something. It would be negligent if IBM didn't persue a Linux strategy.
They've done enough now that recovery is impossible, or very nearly so. The end has come and gone, what you see now is nothing but inertia.
Dunno about anyone else, but Dell will certainly sell you one if you want it. IBM and others probably would too, but I don't feel like checking. Didn't you watch the keynote? They compared a dual G5 to a dual Xeon (an x86 system).
"Apple (If you use Macs): DC 5
Apple (otherwise) : DC 15
Modifiers:
Is switching to linux: -20
Is switching from linux: +15
Is going after Microsoft: -10
_____ vs. SCO : -20
Files a BS patent: +10
Is being investigated by the US government for anti-trust or Fraud: -5"
Makes crappy computers that break a lot: +50
Okay... I have Apple, a DC 5 check. They're not really going to or from Linux. They make OpenFirmware computers and support many of the same APIs, so we'll give them the -20 anyway. They're pretty much the original anti-microsoft, so they get the -10 easy. Doing well so far...
Oh wait. My iBook has needed repairs 6 times in 8 months, accounting for something like 15% of the time I've owned it. I asked them to replace it, but they refused. +50.
I'll have to roll a natural 25 to make this check. I roll a 3. While I have nothing against their IP practices, or their stance towards Microsoft, their computers are too crappy for me to ever consider buying one again, even if it's the best tool for the job.
QNX is a real OS... and they probably got source too.
In an OS course, you want a clean OS so you can stick to the theory. Linux is HUGE in comparison, and it is not a clean design.
It's about actually having a good chance of understanding a significant portion of the OS, not wading through performance hacks.
"However, I will grant that the average /. user knows what they're doing with their data far more than my average user does and is less likely to cause self-inflicted damage."
eh...
I know what I'm doing and I've found some pretty impressive ways to shoot myself in the foot. Right now the knowledge just makes me dangerous. I wouldn't let me anywhere near a production system.
FreeBSD 4.8 doesn't work with my Gigabyte GA-8IG1000MK's ethernet controller... it's supposed to support it with the fxp(4) driver, but it doesn't. I was sad. I don't think it's fixed in 4.9, at least the release notes don't mention it.
Gentoo is acceptable, but I had trouble with it too. "emerge e100" installed a module for the wrong kernel version and broke it completely. I had to add support manually. Easy (about 20 seconds in "make menuconfig"), but annoying. Pretty good otherwise.
I think I'd like to use FreeBSD just because I like BSD, but it's not quite as slick as OpenBSD. I don't think I'd like OpenBSD as a desktop system...
I saw one in Australia a while ago that could only function with the DOS drivers, even in Windows. Really, really cheap brand, I'd never even heard of it before. That was the only one I had seen.
I've seen some weird problems on LG drives though. Specific combinations of hardware have made mine unusable after waking from hiberation on an XP system. Plain ol' suspend doesn't bother it. That drive doesn't have any problems on Linux systems though. But then, I don't use Mandrake...
"Many people don't have the luxury of grabbing a P4 or Opteron as there whims demand. Supporting older hardware also does not slow an OS down. If done properly, it forces you to think about how to do things RIGHT so you don't eat unnecessary cpu cycles."
It's not just for speed. Multiple architechtures can break bad code early.
"By having multiple register sets within a single core, and tagging requests/results, you can avoid the complexity of SMP entirely, while producing the effect of having multiple processors."
No... keeping all the execution units of a CPU busy all the time is hard, and there's a limit to how much you can do. Hyperthreading is perhaps the more elegant solution, but it has its problems too. It tends to thrash the cache with certain algorithms, and the scalability is limited.
Adding a core increases instruction throughput a lot without adding (too much) complexity, and whatever you might think complexity makes things really tough. It's simply cheaper to throw more cores at the problem.
Another thing you don't consider is that signal propogation delays are a very real problem in todays CPUs. That's only going to get worse. Some way to spread out intependant operation must be used to keep things running fast.
um... no
RISC made a lot more sense when you couldn't fit a CPU onto a single die. You couldn't just throw transistors at it like you can now.
The internal architechture isn't and shouldn't be exposed to the outside. If it is, then compatability is broken each time they want to change it, and that simply is not an option. Compilers would never have a chance to get very good.
RISC binaries are a bit bigger, but the internal opcodes x86 chips use are absoloutely huge, like over a hundred bits per instruction. That would fill up the cache in no time, and memory latency and bandwidth would become a bigger problem than it is now.
What's the big deal?
He's an asshole. So is Rieser. So is Stallman. A lot of the people in the open source community are. Who cares? You don't have to deal with him.
"I don't think any of the free *nixes are better or worse as desktop systems."
Eh -- I dunno. Everything targets Linux. Takes a lot less work to get stuff working on it, especially like video cards and stuff. Also, consider the better interactivity on 2.6. Not appropriate for servers yet, but should be fine for a plain old dekstop. OpenBSD can't even take full advantage of a P4 HT.
A firewall system is a different story. Your basic scrap computer will do just fine, as long as it can fit 2 PCI NIC's. It's better for security if it's a seperate box from everything else, and it's cheaper if the computer that stays on all the time can do so on like 20 watts.
"Since I hear only good things about pf"
gah... you don't even want to know. After you've screwed around with it for a few hours it's easier than a hardware router with a web interface. Complex stuff can be done in one line, and while it requires thought, it has a much smaller tendency to bite you in the ass than with the other firewalls I've dealt with, because the language is smart and intuitive.
And it's fast.
I love PF. It beats the crap out of netfilter/ipchains, IPF, and it's almost easier to use than a web interface on a hardware router.
Dunno how happy you'll be with it as a desktop system. My OpenBSD machine lives in the basement crawlspace, and I'm reasonably sure it doesn't have a monitor or keyboard at the moment. I don't know what it's like with a GUI.
I agree...
Authors are free to use whatever license they choose and I am free to not use their software. That said, I find the GPL hypocritical. "This is Free Software, subject to the following conditions.".
It seems to be working, but the BSD license seems to be working too (Before anyone says BSD is dying, consider that things like Apache use a BSD-like license).
I googled around, and looked on Stallman's site and the GNU site and I couldn't find it... do you have any links?
I had forgotten that...
:)
I stand by my assessment though.
Sadly, BSD is getting soft too. Everything on my OpenBSD system keeps working without me having to get my hands dirty. FreeBSD is worse. Don't even get me started on OS X...
NetBSD still has promise though. It's practically unusable. It doesn't even have sudo installed by default. If you're in a masochistic mood, try it out.
OpenSSH, OpenSSL and Apache are all released under BSD-like licenses. By BSD-like, I mean you don't have to distribute the source of derivative works. Dunno about the other stuff, didn't feel like checking. :)
"And, they have to kill BSD as a loophole for source; otherwise, they're in deep doo-doo in court since it seems they didn't realize that what they took is from BSD."
That would be very, very difficult. The rights that SCO is trying to work with are the same ones USL (aka AT&T/Novell) tried to use in the BSD lawsuit, which USL settled when BSD code was found in UNIX without the copyright notice. In the settlement, all the BSD's had to go back to an older code base and rewrite a few files, but according to the settlement, after they did that they were clean.
If SCO goes after them, it'll be useless because they've all still got the clean code, but SCO's still using the dirty code.
As I understand it, the reason you can't do that is because 2.6.x breaks some 2.4.x stuff. Having two root partitions could mitigate that, because you'd have a place to put the different binaries and configuration files.
Any claim SCO had over BSD was explicitly given up when Novell settled the lawsuit with Berkley. If SCO did go after BSD, the BSD code that was incorporated into SysV without a copyright notice would make SCO vulnerable to a countersuit, just like it did in the early 90's.
I believe, though I can't prove it, that Apple was responsible for the memory that went bad. It was 3rd party, and after they replaced the motherboard my iBook would crash daily. Eventually the memory failed and I had to RMA it. Kingston was happy to replace it. The problems started immediately after I got the iBook back. Between the time I replaced the memory and the time the motherboard broke again, it was rock solid. I believe I hadn't rebooted once in the ~45 day period.